{ "best_metric": null, "best_model_checkpoint": null, "epoch": 7.861635220125786, "eval_steps": 500, "global_step": 2500, "is_hyper_param_search": true, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "eval_accuracy": 0.5948387096774194, "eval_loss": 0.2074165940284729, "eval_runtime": 5.1592, "eval_samples_per_second": 600.867, "eval_steps_per_second": 12.599, "step": 318 }, { "epoch": 1.5723270440251573, "grad_norm": 0.538582980632782, "learning_rate": 1.650593990216632e-05, "loss": 0.3336, "step": 500 }, { "epoch": 2.0, "eval_accuracy": 0.8258064516129032, "eval_loss": 0.09868818521499634, "eval_runtime": 5.3751, "eval_samples_per_second": 576.732, "eval_steps_per_second": 12.093, "step": 636 }, { "epoch": 3.0, "eval_accuracy": 0.8829032258064516, "eval_loss": 0.06500214338302612, "eval_runtime": 5.0949, "eval_samples_per_second": 608.447, "eval_steps_per_second": 12.758, "step": 954 }, { "epoch": 3.1446540880503147, "grad_norm": 0.4604446291923523, "learning_rate": 1.3011879804332637e-05, "loss": 0.1147, "step": 1000 }, { "epoch": 4.0, "eval_accuracy": 0.8996774193548387, "eval_loss": 0.049765754491090775, "eval_runtime": 5.2676, "eval_samples_per_second": 588.5, "eval_steps_per_second": 12.34, "step": 1272 }, { "epoch": 4.716981132075472, "grad_norm": 0.31128990650177, "learning_rate": 9.517819706498952e-06, "loss": 0.0724, "step": 1500 }, { "epoch": 5.0, "eval_accuracy": 0.91, "eval_loss": 0.04065703973174095, "eval_runtime": 5.2744, "eval_samples_per_second": 587.744, "eval_steps_per_second": 12.324, "step": 1590 }, { "epoch": 6.0, "eval_accuracy": 0.912258064516129, "eval_loss": 0.03559587895870209, "eval_runtime": 5.1021, "eval_samples_per_second": 607.594, "eval_steps_per_second": 12.74, "step": 1908 }, { "epoch": 6.289308176100629, "grad_norm": 0.269597589969635, "learning_rate": 6.02375960866527e-06, "loss": 0.0567, "step": 2000 }, { "epoch": 7.0, "eval_accuracy": 0.9203225806451613, "eval_loss": 0.032886430621147156, "eval_runtime": 5.44, "eval_samples_per_second": 569.85, "eval_steps_per_second": 11.948, "step": 2226 }, { "epoch": 7.861635220125786, "grad_norm": 0.28186649084091187, "learning_rate": 2.5296995108315863e-06, "loss": 0.0496, "step": 2500 } ], "logging_steps": 500, "max_steps": 2862, "num_input_tokens_seen": 0, "num_train_epochs": 9, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 651155886807636.0, "train_batch_size": 48, "trial_name": null, "trial_params": { "alpha": 0.186668286700706, "num_train_epochs": 9, "temperature": 8 } }